WO2013153848A1 - Dispositif et procédé de traitement d'image - Google Patents

Dispositif et procédé de traitement d'image Download PDF

Info

Publication number
WO2013153848A1
WO2013153848A1 PCT/JP2013/053193 JP2013053193W WO2013153848A1 WO 2013153848 A1 WO2013153848 A1 WO 2013153848A1 JP 2013053193 W JP2013053193 W JP 2013053193W WO 2013153848 A1 WO2013153848 A1 WO 2013153848A1
Authority
WO
WIPO (PCT)
Prior art keywords
section
luminance
image data
pixels
gradation
Prior art date
Application number
PCT/JP2013/053193
Other languages
English (en)
Japanese (ja)
Inventor
貴之 木村
Original Assignee
株式会社デンソー
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 株式会社デンソー filed Critical 株式会社デンソー
Priority to US14/391,707 priority Critical patent/US9396527B2/en
Publication of WO2013153848A1 publication Critical patent/WO2013153848A1/fr

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T5/00—Image enhancement or restoration
    • G06T5/90—Dynamic range modification of images or parts thereof
    • G06T5/92—Dynamic range modification of images or parts thereof based on global image properties
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N1/00—Scanning, transmission or reproduction of documents or the like, e.g. facsimile transmission; Details thereof
    • H04N1/40—Picture signal circuits
    • H04N1/407—Control or modification of tonal gradation or of extreme levels, e.g. background level
    • H04N1/4072—Control or modification of tonal gradation or of extreme levels, e.g. background level dependent on the contents of the original
    • H04N1/4074—Control or modification of tonal gradation or of extreme levels, e.g. background level dependent on the contents of the original using histograms
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T5/00—Image enhancement or restoration
    • G06T5/40—Image enhancement or restoration using histogram techniques
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T5/00—Image enhancement or restoration
    • G06T5/90—Dynamic range modification of images or parts thereof
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/10—Segmentation; Edge detection
    • G06T7/11—Region-based segmentation
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/10—Segmentation; Edge detection
    • G06T7/136—Segmentation; Edge detection involving thresholding
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V20/00—Scenes; Scene-specific elements
    • G06V20/50—Context or environment of the image
    • G06V20/56—Context or environment of the image exterior to a vehicle by using sensors mounted on the vehicle
    • G06V20/588—Recognition of the road, e.g. of lane markings; Recognition of the vehicle driving pattern in relation to the road
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00—Indexing scheme for image analysis or image enhancement
    • G06T2207/20—Special algorithmic details
    • G06T2207/20172—Image enhancement details
    • G06T2207/20208—High dynamic range [HDR] image processing
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00—Indexing scheme for image analysis or image enhancement
    • G06T2207/30—Subject of image; Context of image processing
    • G06T2207/30248—Vehicle exterior or interior
    • G06T2207/30252—Vehicle exterior; Vicinity of vehicle
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00—Indexing scheme for image analysis or image enhancement
    • G06T2207/30—Subject of image; Context of image processing
    • G06T2207/30248—Vehicle exterior or interior
    • G06T2207/30252—Vehicle exterior; Vicinity of vehicle
    • G06T2207/30256—Lane; Road marking
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00—Indexing scheme for image analysis or image enhancement
    • G06T2207/30—Subject of image; Context of image processing
    • G06T2207/30248—Vehicle exterior or interior
    • G06T2207/30252—Vehicle exterior; Vicinity of vehicle
    • G06T2207/30261—Obstacle

Definitions

  • the present invention relates to an image processing apparatus and method.
  • An image recognition device that recognizes a white line, a preceding vehicle, an oncoming vehicle, and the like using image data acquired by an in-vehicle camera is known.
  • the performance of the in-vehicle camera has improved, and the in-vehicle camera may output image data exceeding 8 bits.
  • image data exceeding 8 bits may be converted into 8-bit image data and used for image recognition (see Patent Document 1).
  • the present invention has been made in view of the above points, and an object thereof is to provide an image processing apparatus and method in which the luminance resolution is hardly lowered even when the number of gradations of an image is reduced.
  • the image processing apparatus includes, for each section formed by dividing the luminance range in the image data into a plurality of sections of the image data having the first gradation number by the section pixel number calculation means.
  • the number of pixels having luminance is calculated.
  • the section selection unit selects a section from the plurality of sections based on the number of pixels having the luminance in the section.
  • the image processing apparatus of the present invention uses a gradation setting unit to change the luminance of the pixel having the luminance in the section selected by the section selecting unit to a second floor smaller than the first number of gradations.
  • the luminance is set in the logarithm and the luminance of the pixel having the luminance in the section not selected by the section selection means is set to the minimum value or the maximum value.
  • the image processing apparatus when the image data having the second gradation number is generated from the image data having the first gradation number, not the entire luminance range in the image data having the first gradation number, but the section A gradation of the second gradation number is set for the pixel corresponding to the section selected based on the number of pixels. Therefore, the luminance resolution is high in the second gradation number image.
  • FIG. 1 is a block diagram illustrating a configuration of an image processing device 1.
  • FIG. 3 is a flowchart illustrating processing executed by the image processing apparatus 1. It is a graph showing the relationship between the brightness of the light which injects into the vehicle-mounted camera 101, and the brightness
  • FIG. It is a histogram showing the number (frequency) of the pixel which has each gradation of a brightness
  • the configuration of the image processing device 1 will be described with reference to FIG.
  • the image processing apparatus 1 is an in-vehicle device mounted on a vehicle, receives 10-bit image data obtained by photographing the front of the vehicle from the in-vehicle camera 101, converts a part of the image data into 8-bit image data (floor). The logarithm is dropped) and output to the image recognition apparatus 201.
  • the image processing apparatus 1 includes a RAM 3, a processing shape selection unit 5, a histogram processing unit (section pixel number calculation unit) 7, a section selection unit (section selection unit) 9, and an 8-bit gradation setting unit (gradation setting unit) 11. It has.
  • Each configuration of the image processing apparatus 1 is realized by a known CPU (computer).
  • the RAM 3 temporarily stores 10-bit image data received from the in-vehicle camera 101 and outputs it to the processing shape selection unit 5.
  • the processing shape selection unit 5, the histogram processing unit 7, the section selection unit 9, and the 8-bit gradation setting unit 11 respectively perform processing shape selection processing, histogram processing, section selection processing, and 8-bit gradation setting processing described later. Execute to create 8-bit image data.
  • the 8-bit gradation setting unit 11 outputs the created 8-bit image data to the image recognition apparatus 201.
  • the image recognition apparatus 201 temporarily stores 8-bit image data in the RAM 203, and executes image recognition processing using the 8-bit image data.
  • a white line ahead of the vehicle, a tail lamp of a preceding vehicle, a head lamp of an oncoming vehicle, and the like are recognized by a known image recognition technique.
  • the recognition results are Lane Departure Warning, Lane Keeping (Lane Keeping) Assist, Forward Collision Waning, Pre-Crush Safety, Auto Emergency Brake, Auto High Beam, etc.
  • step 1 of FIG. 2 10-bit image data is received from the in-vehicle camera 101 and stored in the RAM 3.
  • Each pixel of 10-bit image data has the number of gradations corresponding to 10 bits, that is, the number of gradations of 2 10 (the first number of gradations) in terms of luminance.
  • this image data is the luminance of the pixels constituting the image data (vertical axis in FIG. 3) with respect to the amount of change in the brightness of light incident on the in-vehicle camera 101 (horizontal axis in FIG. 3).
  • the rate of change (the slope of the graph in FIG. 3) is dynamic range image data that differs for each luminance region. That is, the above-mentioned inclination is greatest in the region from Y 0 (minimum value) to Y 1 , medium in the region from Y 1 to Y 2 , and most in the region from Y 2 to Y 3 (maximum value). small.
  • Y 0 , Y 1 , Y 2, and Y 3 are all luminance (gradation) expressed as a power of 2.
  • the processing shape selection unit 5 selects a portion to be processed (hereinafter referred to as “processing portion”) from the 10-bit image data.
  • the processing portion may be one of the lines 307 along the water direction across the white lines 303 and 305 in front of the vehicle in the 10-bit image data 301.
  • a rectangular area (an area where a preceding vehicle or an oncoming vehicle may exist) 309 in front of the vehicle may be included in the 10-bit image data 301.
  • step 3 the histogram processing unit 7 first acquires the luminance of each pixel present in the processing portion selected in step 2. Next, in the processing portion, the number of pixels having luminance of Y 0 , the number of pixels having luminance one level higher than Y 0 , the number of pixels having luminance two levels higher than Y 0 ,... Y The number of pixels having a luminance of 3 is obtained. As a result, a histogram representing the number (frequency) of pixels having each gradation of luminance shown in FIG. 6 is obtained.
  • the histogram processing unit 7 sets the luminance range (Y 0 to Y 3 ) of the 10-bit image data to the section R 01 from Y 0 to Y 1 , the section R 12 from Y 1 to Y 2 , and Y 2. divided into a section R 23 of ⁇ Y 3. Then, using the histogram of FIG. 6 obtained as described above, the pixel number N 01 which is the number of pixels having the luminance in the interval R 01 in the processing portion, and the pixel having the luminance in the interval R 12 in the processing portion. calculated interval pixel number N 12 is the number, and the processing portion of the section number of pixels N 23 is the number of pixels having a brightness in the interval R 23, respectively.
  • the section selection unit 9 selects a part or all of the sections from the sections R 01 , R 12 , and R 23 based on the number of section pixels N 01 , N 12 , and N 23 as follows. To do. The maximum value among the number of section pixels N 01 , N 12 , and N 23 is defined as N max . Then, it is determined whether or not the value of N max / N all is greater than 0.8 (threshold value). Here, N all is the total of the number of section pixels N 01 , N 12 , and N 23 .
  • the 8-bit gradation setting unit 11 performs an 8-bit gradation setting process for each pixel in the processing portion as follows. First, when the section R 23 is selected in the step 4, as shown in FIG. 7, the pixels of the processing portion having the brightness in the section R 23 have the brightness using the entire range of 8-bit gradation. Is set. At this time, in each pixel, there is a linear relationship between the luminance in the 10-bit image data and the newly set 8-bit gradation. On the other hand, the lowest luminance in the 8-bit gradation is set for the pixels of the processing portion having the luminance in the sections R 01 and R 12 not selected in the step 4.
  • the pixels of the processing portion having the brightness in the section R 12 have the brightness using the entire range of 8-bit gradation. Is set. At this time, in each pixel, there is a linear relationship between the luminance in the 10-bit image data and the newly set 8-bit gradation.
  • the lowest luminance in the 8-bit gradation is set in the pixels of the processing portion having the luminance in the section R 01 not selected in the step 4, and the pixel in the section R 23 not selected in the step 4 is set.
  • the highest luminance in the 8-bit gradation is set in the pixel of the processing portion having the luminance of.
  • the pixels of the processing portion having the brightness in the section R 01 have the brightness using the entire range of 8-bit gradation. Is set. At this time, in each pixel, there is a linear relationship between the luminance in the 10-bit image data and the newly set 8-bit gradation. On the other hand, the highest luminance in the 8-bit gradation is set to the pixels of the processing portion having the luminance in the sections R 12 and R 23 not selected in the step 4.
  • the 8-bit gradation is applied to the pixels having luminance in all the sections.
  • the brightness is set using the entire range. Also in this case, in each pixel, there is a linear relationship between the luminance in the 10-bit image data and the newly set 8-bit gradation.
  • 8-bit image data is created by setting the luminance at the 8-bit gradation for each pixel in the processing portion.
  • the 8-bit image data has a gradation number of 2 8 (second gradation number).
  • step 6 the 8-bit gradation setting unit 11 outputs the 8-bit image data created in step 5 to the image recognition apparatus 201. 3. Effects produced by the image processing apparatus 1 (1) When the image processing apparatus 1 creates 8-bit image data from 10-bit image data, the number of corresponding pixels is not the entire range of luminance in the 10-bit image data. A section having a large number of images (an important section in image recognition) is selected, and an 8-bit gradation is set for pixels corresponding to the section. Therefore, the luminance resolution is high in an 8-bit image.
  • the image processing apparatus 1 determines that all of the pixels in all the sections are 8 if there is no section in which the number of corresponding pixels is significantly high in the three sections (sections R 01 , R 12 , and R 23 ) relating to luminance. Sets the bit gradation. Therefore, even when there is no section in which the number of corresponding pixels is remarkably high, 8-bit image data can be normally created. (3) The image processing apparatus 1 performs processing on a processing portion (a line 307 in FIG. 4 and a region 309 in FIG. 5) which is a part of 10-bit image data. Therefore, processing can be performed quickly. (4) Both the 10-bit image data and the 8-bit image data created by the image processing apparatus 1 based on the 10-bit image data are dynamic range image data. Therefore, an image corresponding to a wide range regarding brightness can be created.
  • image data input to the image processing apparatus 1 is Instead of 10-bit image data, image data of other gradation numbers (for example, 12-bit, 14-bit, 16-bit image data, etc.) may be used. Further, the image data input to the image processing apparatus 1 may not be dynamic range image data but may be image data having linear characteristics of brightness and luminance.
  • the number of dividing the luminance range is not limited to 3, and may be a plurality other than 3 (for example, 2, 4, 5, 6,).
  • the method may be other methods.
  • the frequency (for example, (N 01 + N 12 ) / N all or (N 12 + N 23 ) / N corresponding to two consecutive sections of the sections R 01 , R 12 , and R 23. all ) is greater than the first threshold, and if the frequency of the number of pixels corresponding to one of the two consecutive sections is greater than the second threshold, that section is selected. be able to.
  • N 01 / N all , N 12 / N all , and N 23 / N all may be calculated, and each of them may be compared with a threshold value, and all those larger than the threshold value may be selected. In this case, there are only one section to be selected, two cases, and three cases.
  • Step 5 another threshold value may be used instead of the threshold value 0.8 used in Step 4 above.
  • another gradation for example, 6 bits, 4 bits, the number obtained by dividing the first gradation number by 2, etc.
  • another gradation for example, 6 bits, 4 bits, the number obtained by dividing the first gradation number by 2, etc.
  • the luminance of the pixel corresponding to the section not selected in Step 5 may be either the lowest value or the highest value in the 8-bit gradation. Further, the image processing apparatus 1 may perform processing on the entire 10-bit image data.
  • DESCRIPTION OF SYMBOLS 1 ... Image processing apparatus, 3 ... RAM, 5 ... Processing shape selection part, 7 ... Histogram processing part, 9 ... Section selection part, 11 ... 8-bit gradation setting part, DESCRIPTION OF SYMBOLS 101 ... Car-mounted camera, 201 ... Image recognition apparatus, 301 ... 10-bit image data, 303 ... White line, 307 ... Line, 309 ... Area

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Multimedia (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Signal Processing (AREA)
  • Image Processing (AREA)
  • Facsimile Image Signal Circuits (AREA)

Abstract

La présente invention concerne un dispositif et un procédé de traitement d'image comprenant : un moyen de calcul de nombre de pixels d'intervalles (7, S3) pour diviser la plage de luminosité de données d'image (301) ayant une première échelle de gris en une pluralité d'intervalles, et calculer le nombre de pixels ayant des valeurs de luminosité dans chacun des intervalles ; un moyen de sélection d'intervalle (9, S4) pour sélectionner, en fonction du nombre de pixels ayant des valeurs de luminosité pour chacun des intervalles, un intervalle parmi la pluralité d'intervalles ; et un moyen de réglage d'échelle de gris (11, S5) pour régler les valeurs de luminosité des pixels ayant les valeurs de luminosité dans l'intervalle sélectionné par le moyen de sélection d'intervalle sur une seconde échelle de gris inférieure à la première échelle de gris, et régler les valeurs de luminosité des pixels ayant les valeurs de luminosité dans un intervalle non sélectionné par le moyen de sélection d'intervalle sur une valeur minimale ou sur une valeur maximale.
PCT/JP2013/053193 2012-04-13 2013-02-12 Dispositif et procédé de traitement d'image WO2013153848A1 (fr)

Priority Applications (1)

Application Number Priority Date Filing Date Title
US14/391,707 US9396527B2 (en) 2012-04-13 2013-02-12 Image processing device and method

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
JP2012092198A JP5919988B2 (ja) 2012-04-13 2012-04-13 画像処理装置
JP2012-092198 2012-04-13

Publications (1)

Publication Number Publication Date
WO2013153848A1 true WO2013153848A1 (fr) 2013-10-17

Family

ID=49327428

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/JP2013/053193 WO2013153848A1 (fr) 2012-04-13 2013-02-12 Dispositif et procédé de traitement d'image

Country Status (3)

Country Link
US (1) US9396527B2 (fr)
JP (1) JP5919988B2 (fr)
WO (1) WO2013153848A1 (fr)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN107665480A (zh) * 2016-07-28 2018-02-06 佳能株式会社 图像处理装置、其控制方法、显示装置及存储介质

Families Citing this family (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN109754410B (zh) * 2019-01-03 2020-12-11 北京化工大学 一种基于机器视觉的铁路车辆车厢计数方法

Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2006098360A1 (fr) * 2005-03-15 2006-09-21 Omron Corporation Processeur d'images, procede et systeme de traitement d'images, programme et support d'enregistrement
JP2010057080A (ja) * 2008-08-29 2010-03-11 Fujitsu Frontech Ltd 画像処理装置、画像処理方法および画像処理プログラム
JP2010278724A (ja) * 2009-05-28 2010-12-09 Olympus Corp 画像処理装置、画像処理方法及び画像処理プログラム

Family Cites Families (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US8599257B2 (en) * 2006-08-18 2013-12-03 Nec Corporation Vehicle detection device, vehicle detection method, and vehicle detection program
JP5073548B2 (ja) * 2008-03-27 2012-11-14 富士重工業株式会社 車両用環境認識装置および先行車追従制御システム
JP4666049B2 (ja) * 2008-10-17 2011-04-06 株式会社デンソー 光源識別装置、光源識別プログラム、車両検出装置、およびライト制御装置

Patent Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2006098360A1 (fr) * 2005-03-15 2006-09-21 Omron Corporation Processeur d'images, procede et systeme de traitement d'images, programme et support d'enregistrement
JP2010057080A (ja) * 2008-08-29 2010-03-11 Fujitsu Frontech Ltd 画像処理装置、画像処理方法および画像処理プログラム
JP2010278724A (ja) * 2009-05-28 2010-12-09 Olympus Corp 画像処理装置、画像処理方法及び画像処理プログラム

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN107665480A (zh) * 2016-07-28 2018-02-06 佳能株式会社 图像处理装置、其控制方法、显示装置及存储介质

Also Published As

Publication number Publication date
US20150110397A1 (en) 2015-04-23
US9396527B2 (en) 2016-07-19
JP5919988B2 (ja) 2016-05-18
JP2013222261A (ja) 2013-10-28

Similar Documents

Publication Publication Date Title
JP5889431B2 (ja) 画像処理装置及び画像処理方法
JP4795473B2 (ja) 画像処理装置及びその制御方法
US9466097B2 (en) Apparatus and method for removing fog in image
KR100901353B1 (ko) 영상 처리 장치 및 그 방법
JP6064776B2 (ja) 画像処理装置及び画像処理方法
JP2006197584A (ja) 画像のｒｇｂ情報を利用したブラック／ホワイト拡張システム
KR20140134374A (ko) 다수 영상의 밝기 균일화 방법
JPWO2008139565A1 (ja) 暗視装置
WO2013153848A1 (fr) Dispositif et procédé de traitement d'image
CN112508883A (zh) 一种列车车轮踏面锥形孔检测的自适应缺陷识别方法
JP5089797B2 (ja) 画像処理装置及びその制御方法
US9727794B2 (en) Image processing apparatus and lane partition line recognition system including the same
KR20130133371A (ko) 적응적 가중치 예측을 이용한 영상 처리 방법
JP5202749B1 (ja) 画像処理方法
JP7505596B2 (ja) 画像処理装置、画像処理方法、及び画像処理プログラム
KR101497933B1 (ko) 클러스터링 기법을 이용한 다수 영상의 합성 시스템 및 방법
JP6589741B2 (ja) 画像処理装置
JP2007259148A (ja) ヒストグラムプロジェクション処理用度数閾値設定装置、方法、及びそのプログラムを記録した記録媒体。
EP4390836A1 (fr) Amélioration d'image ciblée lors de l'adressage de dégradations provoquées par des conditions environnementales
JP2006331163A (ja) 画像処理装置
WO2012048661A1 (fr) Procédé et dispositif de tramage
US20210150250A1 (en) Image processing device and image processing method
US20230230216A1 (en) Image processing apparatus, image processing method, image processing system, and storage medium
JP7145651B2 (ja) 画像処理装置及びプログラム
JP2018022294A (ja) 画像処理装置とその制御方法、及びプログラム

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 13774940

Country of ref document: EP

Kind code of ref document: A1

WWE Wipo information: entry into national phase

Ref document number: 14391707

Country of ref document: US

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 13774940

Country of ref document: EP

Kind code of ref document: A1